The term mid-flexion instability has entered the orthopaedic literature as a concept, but has not been confirmed as a distinct clinical entity. The term is used freely, sometimes as a synonym for flexion instability. However, the terms need to be clearly separated. A cadaver study published in 1990 associated joint line elevation with decreased stability at many angles of flexion, but that model was not typical of clinical scenarios. The literature is considered and it is proposed that the more common entity of an uncorrected flexion contracture after a measured resection arthroplasty technique is more likely to produce clinical findings that suggest instability mid-flexion.

It is proposed that the clinical scenario encountered is generalised instability, with the appearance of stability in full extension from tight posterior structures.

This paper seeks to clarify whether mid-flexion instability exists as an entity distinct from other commonly recognised forms of instability.

Cite this article: Bone Joint J 2016;98-B(1 Suppl A):84–8.

We analysed the serum C-reactive protein level, synovial fluid obtained by joint aspiration and five synovial biopsies from 145 knee replacements prior to revision to assess the value of these parameters in diagnosing late peri-prosthetic infection. Five further synovial biopsies were used for histological analysis. Samples were also obtained during the revision and incubated and analysed in an identical manner for 14 days. A total of 40 total knee replacements were found to be infected (prevalence 27.6%). The aspiration technique had a sensitivity of 72.5% (95% confidence interval (CI) 58.7 to 86.3), a specificity of 95.2% (95% CI 91.2 to 99.2), a positive predictive value of 85.3% (95% CI 73.4 to 100), a negative predictive value of 90.1% (95% CI 84.5 to 95.7) and an accuracy of 89%. The biopsy technique had a sensitivity of 100%, a specificity of 98.1% (95% CI 95.5 to 100), a positive predictive value of 95.2% (95% CI 88.8 to 100), a negative predictive value of 100% and an accuracy of 98.6%. C-reactive protein with a cut-off-point of 13.5 mg/l had a sensitivity of 72.5% (95% CI 58.7 to 86.3), a specificity of 80.9% (95% CI 73.4 to 88.4), a positive predictive value of 59.2% (95% CI 45.4 to 73.0), a negative predictive value of 88.5% (95% 81.0 to 96.0 CI) and an accuracy of 78.1%. We found that biopsy was superior to joint aspiration and C-reactive protein in the diagnosis of late peri-prosthetic infection of total knee replacements

The role of arthroscopy in the treatment of soft-tissue injuries associated with proximal tibial fractures remains debatable. Our hypothesis was that MRI over-diagnoses clinically relevant associated soft-tissue injuries. This prospective study involved 50 consecutive patients who underwent surgical treatment for a split-depression fracture of the lateral tibial condyle (AO/OTA type B3.1). The mean age of patients was 50 years (23 to 86) and 27 (54%) were female. All patients had MRI and arthroscopy. Arthroscopy identified 12 tears of the lateral meniscus, including eight bucket-handle tears that were sutured and four that were resected, as well as six tears of the medial meniscus, of which five were resected. Lateral meniscal injuries were diagnosed on MRI in four of 12 patients, yielding an overall sensitivity of 33% (95% confidence interval (CI) 11 to 65). Specificity was 76% (95% CI 59 to 88), with nine tears diagnosed among 38 menisci that did not contain a tear. MRI identified medial meniscal injuries in four of six patients, yielding an overall sensitivity of 67% (95% CI 24 to 94). Specificity was 66% (95% CI 50 to 79), with 15 tears diagnosed in 44 menisci that did not contain tears.

MRI appears to offer only a marginal benefit as the specificity and sensitivity for diagnosing meniscal injuries are poor in patients with a fracture. There were fewer arthroscopically-confirmed associated lesions than reported previously in MRI studies.

Cite this article: Bone Joint J 2014;96-B:1631–6.

Aims. To identify unanswered questions about the prevention, diagnosis, treatment, and rehabilitation and delivery of care of first-time soft-tissue knee injuries (ligament injuries, patella dislocations, meniscal injuries, and articular cartilage) in children (aged 12 years and older) and adults. Methods. The James Lind Alliance (JLA) methodology for Priority Setting Partnerships was followed. An initial survey invited patients and healthcare professionals from the UK to submit any uncertainties regarding soft-tissue knee injury prevention, diagnosis, treatment, and rehabilitation and delivery of care. Over 1,000 questions were received. From these, 74 questions (identifying common concerns) were formulated and checked against the best available evidence. An interim survey was then conducted and 27 questions were taken forward to the final workshop, held in January 2023, where they were discussed, ranked, and scored in multiple rounds of prioritization. This was conducted by healthcare professionals, patients, and carers. Results. The top ten included questions regarding prevention, diagnosis, treatment, and rehabilitation. The number one question was, ‘How urgently do soft-tissue knee injuries need to be treated for the best outcome?’. This reflects the concerns of patients, carers, and the wider multidisciplinary team. Conclusion. This validated process has generated ten important priorities for future soft-tissue knee injury research. These have been submitted to the National Institute for Health and Care Research. All 27 questions in the final workshop have been published on the JLA website. Cite this article: Bone Joint J 2024;106-B(3):232–239

Aims. The aim of this study was to establish consensus statements on the diagnosis, nonoperative management, and indications, if any, for medial patellofemoral complex (MPFC) repair in patients with patellar instability, using the modified Delphi approach. Methods. A total of 60 surgeons from 11 countries were invited to develop consensus statements based on their expertise in this area. They were assigned to one of seven working groups defined by subtopics of interest within patellar instability. Consensus was defined as achieving between 80% and 89% agreement, strong consensus was defined as between 90% and 99% agreement, and 100% agreement was considered to be unanimous. Results. Of 27 questions and statements on patellar instability, three achieved unanimous consensus, 14 achieved strong consensus, five achieved consensus, and five did not achieve consensus. Conclusion. The statements that reached unanimous consensus were that an assessment of physeal status is critical for paediatric patients with patellar instability. There was also unanimous consensus on early mobilization and resistance training following nonoperative management once there is no apprehension. The statements that did not achieve consensus were on the importance of immobilization of the knee, the use of orthobiologics in nonoperative management, the indications for MPFC repair, and whether a vastus medialis oblique advancement should be performed. Cite this article: Bone Joint J 2023;105-B(12):1259–1264

Aims. This study aims to determine difference in annual rate of early-onset (≤ 90 days) deep surgical site infection (SSI) following primary total knee arthroplasty (TKA) for osteoarthritis, and to identify risk factors that may be associated with infection. Methods. This is a retrospective population-based cohort study using prospectively collected patient-level data between 1 January 2013 and 1 March 2020. The diagnosis of deep SSI was defined as per the Centers for Disease Control/National Healthcare Safety Network criteria. The Mann-Kendall Trend test was used to detect monotonic trends in annual rates of early-onset deep SSI over time. Multiple logistic regression was used to analyze the effect of different patient, surgical, and healthcare setting factors on the risk of developing a deep SSI within 90 days from surgery for patients with complete data. We also report 90-day mortality. Results. A total of 39,038 patients underwent primary TKA for osteoarthritis during the study period. Of these, 275 patients developed a deep SSI within 90 days of surgery, representing a cumulative incidence of 0.7%. The annual infection rate did not significantly decrease over the seven-year study period (p = 0.162). Overall, 13,885 (35.5%) cases were excluded from the risk analysis due to missing data. Risk factors associated with early-onset deep SSI included male sex, American Society of Anesthesiologists grade ≥ 3, blood transfusion, acute length of stay, and surgeon volume < 30 TKAs/year. Early-onset deep SSI was not associated with increased 90-day mortality. Conclusion. This study establishes a reliable baseline infection rate for early-onset deep SSI after TKA for osteoarthritis using robust Infection Prevention and Control surveillance data, and identifies several potentially modifiable risk factors. Cite this article: Bone Joint J 2023;105-B(9):971–976

Aims. The diagnosis of periprosthetic joint infection (PJI) continues to present a significant clinical challenge. New biomarkers have been proposed to support clinical decision-making; among them, synovial fluid alpha-defensin has gained interest. Current research methodology suggests reference methods are needed to establish solid evidence for use of the test. This prospective study aims to evaluate the diagnostic accuracy of high-performance liquid chromatography coupled with the mass spectrometry (LC-MS) method to detect alpha-defensin in synovial fluid. Methods. Between October 2017 and September 2019, we collected synovial fluid samples from patients scheduled to undergo revision surgery for painful total knee arthroplasty (TKA). The International Consensus Meeting criteria were used to classify 33 PJIs and 92 aseptic joints. LC-MS assay was performed to measure alpha-defensin in synovial fluid of all included patients. Sensitivity, specificity, positive predictive value, negative predictive value, and the area under the receiver operating characteristic curve (AUC) were calculated to define the test diagnostic accuracy. Results. The AUC was 0.99 (95% confidence interval (CI) 0.98 to 1.00). Receiver operating characteristic (ROC) analysis showed that the optimal cut-off value of synovial fluid alpha-defensin was 1.0 μg/l. The sensitivity of alpha-defensin was 100% (95% CI 96 to 100), the specificity was 97% (95% CI 90 to 98), the positive predictive value was 89.2% (95% CI 82 to 94), and negative predictive value was 100% (95% CI 96 to 100). ROC analysis demonstrated an AUC of 0.99 (95% CI 0.98 to 1.0). Conclusion. The present study confirms the utility of alpha-defensin in the synovial fluid in patients with painful TKA to select cases of PJI. Since LC-MS is still a time-consuming technology and is available in highly specialized laboratories, further translational research studies are needed to take this evidence into routine procedures and promote a new diagnostic approach. Cite this article: Bone Joint J 2022;104-B(9):1047–1051

Aims. The aim of this study was to assess factors associated with the estimated lifetime risk of revision surgery after primary knee arthroplasty (KA). Methods. All patients from the Scottish Arthroplasty Project dataset undergoing primary KA during the period 1 January 1998 to 31 December 2019 were included. The cumulative incidence function for revision and death was calculated up to 20 years. Adjusted analyses used cause-specific Cox regression modelling to determine the influence of patient factors. The lifetime risk was calculated as a percentage for patients aged between 45 and 99 years using multiple-decrement life table methodology. Results. The estimated lifetime risk of revision ranged between 32.7% (95% confidence interval (CI) 22.6 to 47.3) for patients aged 45 to 49 years and 0.6% (95% CI 0.1 to 4.5) for patients aged over 90 years. At 20 years, the overall cumulative incidence of revision (6.8% (95% CI 6.6 to 7.0)) was significantly less than that of death (66.3% (95% CI 65.4 to 67.1)). Adjusted analyses demonstrated converse effect of increasing age on risk of revision (hazard ratio (HR) 0.5 (95% CI 0.5 to 0.6)) and death (HR 3.6 (95% CI 3.4 to 3.7)). Male sex was associated with increased risks of revision (HR 1.1 (95% CI 1.1 to 1.2); p < 0.001) and death (HR 1.4 (95% CI 1.3 to 1.4); p < 0.001). Compared to patients undergoing primary KA for osteoarthritis, patients with inflammatory arthropathy had a higher risk of death (HR 1.7 (95% CI 1.7 to 1.8); p < 0.001), but were less likely to be revised (HR 0.9 (95% CI 0.7 to 1.0); p < 0.001). Patients with a greater number of comorbidities (HR 1.4 (95% CI 1.3 to 1.4)) and greater levels of socioeconomic deprivation (HR 1.4 (95% CI 1.4 to 1.5)) were at increased risk of death, but neither increased the risk of revision. Conclusion. The estimated lifetime risk of revision KA varied depending on patient sex, age, and underlying diagnosis. Patients aged between 45 and 49 years had a one in three risk of undergoing revision surgery within their lifetime, which decreased with age to one in 159 in those aged 90 years or more. Cite this article: Bone Joint J 2022;104-B(12):1313–1322

Aims. Fungal periprosthetic joint infections (PJIs) are rare, but their diagnosis and treatment are highly challenging. The purpose of this study was to investigate the clinical outcomes of patients with fungal PJIs treated with two-stage exchange knee arthroplasty combined with prolonged antifungal therapy. Methods. We reviewed our institutional joint arthroplasty database and identified 41 patients diagnosed with fungal PJIs and treated with two-stage exchange arthroplasty after primary total knee arthroplasty (TKA) between January 2001 and December 2020, and compared them with those who had non-fungal PJIs during the same period. After propensity score matching based on age, sex, BMI, American Society of Anesthesiologists grade, and Charlson Comorbidity Index, 40 patients in each group were successfully matched. The surgical and antimicrobial treatment, patient demographic and clinical characteristics, recurrent infections, survival rates, and relevant risk factors that affected joint survivorship were analyzed. We defined treatment success as a well-functioning arthroplasty without any signs of a PJI, and without antimicrobial suppression, at a minimum follow-up of two years from the time of reimplantation. Results. The fungal PJI group demonstrated a significantly worse treatment success rate at the final follow-up than the non-fungal PJI group (65.0% (26/40) vs 85.0% (34/40); p < 0.001). The mean prosthesis-free interval was longer in the fungal PJI group than in the non-fungal PJI group (6.7 weeks (SD 5.8) vs 4.1 weeks (SD 2.5); p = 0.020). The rate of survivorship free from reinfection was worse in the fungal PJI group (83.4% (95% confidence interval (CI) 64.1 to 92.9) at one year and 76.4% (95% CI 52.4 to 89.4) at two years) than in the non-fungal PJI group (97.4% (95% CI 82.7 to 99.6) at one year and 90.3% (95% CI 72.2 to 96.9) at two years), but the differences were not significant (p = 0.270). Cox proportional hazard regression analysis identified the duration of the prosthesis-free interval as a potential risk factor for failure (hazard ratio 1.128 (95% CI 1.003 to 1.268); p = 0.043). Conclusion. Fungal PJIs had a lower treatment success rate than non-fungal PJIs despite two-stage revision arthroplasty and appropriate antifungal treatment. Our findings highlight the need for further developments in treating fungal PJIs. Cite this article: Bone Joint J 2023;105-B(12):1286–1293

Aims. Stiffness is a common complication after total knee arthroplasty (TKA). Pathogenesis is not understood, treatment options are limited, and diagnosis is challenging. The aim of this study was to investigate if MRI can be used to visualize intra-articular scarring in patients with stiff, painful knee arthroplasties. Methods. Well-functioning primary TKAs (n = 11), failed non-fibrotic TKAs (n = 5), and patients with a clinical diagnosis of fibrosis. 1. (n = 8) underwent an MRI scan with advanced metal suppression (Slice Encoding for Metal Artefact Correction, SEMAC) with gadolinium contrast. Fibrotic tissue (low intensity on T1 and T2, low-moderate post-contrast enhancement) was quantified (presence and tissue thickness) in six compartments: supra/infrapatella, medial/lateral gutters, and posterior medial/lateral. Results. Fibrotic tissue was identified in all patients studied. However, tissue was significantly thicker in fibrotic patients (4.4 mm ± 0.2 mm) versus non-fibrotic (2.5 mm ± 0.4 mm) and normal TKAs (1.9 mm ± 0.2 mm, p = < 0.05). Significant (> 4 mm thick) tissue was seen in 26/48 (54%) of compartments examined in the fibrotic group, compared with 17/30 (57%) non-fibrotic, and 10/66 (15%) normal TKAs. Although revision surgery did improve range of movement (ROM) in all fibrotic patients, clinically significant restriction remained post-surgery. Conclusion. Stiff TKAs contain intra-articular fibrotic tissue that is identifiable by MRI. Studies should evaluate whether MRI is useful for surgical planning of debridement, and as a non-invasive measurement tool following interventions for stiffness caused by fibrosis. Revision for stiffness can improve ROM, but outcomes are sub-optimal and new treatments are required. Cite this article: Bone Joint J 2020;102-B(10):1331–1340

Aims. A limited number of investigations with conflicting results have described perivascular lymphocytic infiltration (PVLI) in the setting of total knee arthroplasty (TKA). The purpose of this study was to determine if PVLI found in TKAs at the time of aseptic revision surgery was associated with worse clinical outcomes and survivorship. Methods. A retrospective review was conducted on 617 patients who underwent aseptic TKA revision who had histological analysis for PVLI at the time of surgery. Clinical and radiological data were obtained pre- and postoperatively, six weeks postoperatively, and then every year thereafter. Results. Within this cohort, 118 patients (19.1%) were found to have PVLI on histological analysis. Re-revision was performed on 83 patients (13.4%) with no significant differences in all-cause or aseptic revisions between groups. A higher incidence of PVLI was noted in female patients (p = 0.037). There was no significant difference in improvement in the range of motion (p = 0.536), or improvement of KSC (p = 0.66), KSP (p = 0.61), or KSF (p = 0.3) clinical outcome scores between PVLI and no PVLI sub-groups. There was a higher incidence of a preoperative diagnosis of pain in the PVLI group compared with patients without PVLI (p = 0.002) present. Conclusion. PVLI found on large-scale histological analysis in TKAs at aseptic revision surgery was not associated with worse clinical outcomes or rates of re-revision. Cite this article: Bone Joint J 2021;103-B(6 Supple A):145–149

Aims. This study aimed to determine outcomes of isolated tibial insert exchange (ITIE) during revision total knee arthroplasty (TKA). Methods. From 1985 to 2016, 270 ITIEs were performed at one institution for instability (55%, n = 148), polyethylene wear (39%, n = 105), insert fracture/dissociation (5%, n = 14), or stiffness (1%, n = 3). Patients with component loosening, implant malposition, infection, and extensor mechanism problems were excluded. Results. Survivorship free of any re-revision was 68% at ten years. For the indication of insert wear, survivorship free of any re-revision at ten years was 74%. Re-revisions were more frequent for index diagnoses other than wear (hazard ratio (HR) 1.9; p = 0.013), with ten-year survivorships of 69% for instability and 37% for insert fracture/dissociation. Following ITIE for wear, the most common reason for re-revision was aseptic loosening (33%, n = 7). For other indications, the most common reason for re-revision was recurrence of the original diagnosis. Mean Knee Society Scores improved from 54 (0 to 94) preoperatively to 77 (38 to 94) at ten years. Conclusion. After ITIE, the risk and reasons for re-revision correlated with preoperative indications. The best results were for polyethylene wear. For other diagnoses, the re-revision rate was higher and the failure mode was most commonly recurrence of the original indication for the revision TKA. Cite this article: Bone Joint J 2021;103-B(6):1103–1110

Aims. Tibial cones are often utilized in revision total knee arthroplasty (TKA) with metaphyseal defects. Because there are few studies evaluating mid-term outcomes with a sufficient cohort, the purpose of this study was to evaluate tibial cone survival and complications in revision TKAs with tibial cones at minimum follow-up of five years. Methods. A retrospective review was completed from September 2006 to March 2015, evaluating 67 revision TKAs (64 patients) that received one specific porous tibial cone during revision TKA. The final cohort was composed of 62 knees (59 patients) with five years of clinical follow-up or reoperation. The mean clinical follow-up of the TKAs with minimum five-year clinical follow-up was 7.6 years (5.0 to 13.3). Survivorship analysis was performed with the endpoints of tibial cone revision for aseptic loosening, tibial cone revision for any reason, and reoperation. We also evaluated periprosthetic joint infection (PJI), risk factors for failure, and performed a radiological review. Results. The rate of cone revision for aseptic loosening was 6.5%, with an eight-year survival of 95%. Significant bone loss (Anderson Orthopaedic Research Institute grade 3) was associated with cone revision for aseptic loosening (p = 0.002). The rate of cone revision for any reason was 17.7%, with an eight-year survival of 84%. Sixteen percent of knees developed PJI following revision. A pre-revision diagnosis of reimplantation as part of a two-stage exchange protocol for infection was associated with both PJI (p < 0.001) and tibial cone revision (p = 0.001). Conclusion. Mid-term results of tibial cones showed a survivorship free of cone revision for aseptic loosening of 95%. Patients with significant bone loss were more likely to have re-revision for tibial cone failure. Infection was common, and patients receiving cones at reimplantation were more likely to develop PJI and undergo cone revision. Cite this article: Bone Joint J 2021;103-B(6 Supple A):158–164

Aims. It remains difficult to diagnose early postoperative periprosthetic joint infection (PJI) following total knee arthroplasty (TKA). We aimed to validate the optimal cutoff values of ESR, CRP, and synovial fluid analysis for detecting early postoperative PJI in a large series of primary TKAs. Methods. We retrospectively identified 27,066 primary TKAs performed between 2000 and 2019. Within 12 weeks, 169 patients (170 TKAs) had an aspiration. The patients were divided into two groups: those evaluated ≤ six weeks, or between six and 12 weeks postoperatively. The 2011 Musculoskeletal Infection Society (MSIS) criteria for PJI diagnosis in 22 TKAs. The mean follow-up was five years (two months to 17 years). The results were compared using medians and Mann-Whitney U tests and thresholds were analyzed using receiver operator characteristic curves. Results. Within six weeks, the median CRP (101 mg/l vs 35 mg/l; p = 0.011), synovial WBCs (58,295 cells/μl vs 2,121 cells/μl; p ≤ 0.001), percentage of synovial neutrophils (91% vs 71% (p < 0.001), and absolute synovial neutrophil count (ANC) (50,748 cells/μl vs 1,386 cells/μl (p < 0.001) were significantly higher in infected TKAs. Between six and 12 weeks, the median CRP (85 mg/l vs 5 mg/l (p < 0.001)), ESR (33 mm/hr vs 14 mm/hr (p = 0.015)), synovial WBCs (62,247 cells/μl vs 620 cells/μl (p < 0.001)), percentage of synovial neutrophils (93% vs 54% (p < 0.001)), and ANC (55,911 cells/μl vs 326 cells/μl (p < 0.001)) were also significantly higher in infected TKAs. Optimal thresholds at ≤ six weeks were: CRP ≥ 82 mg/l (sensitivity 70%, specificity 77%), synovial WBCs ≥ 8,676 cells/μl (83%, 90%), percentage of synovial neutrophils ≥ 88% (67%, 78%), and ANC ≥ 8,346 cells/μl (83%, 91%). Between six and 12 weeks, thresholds were: CRP ≥ 34 mg/l (90%, 93%), synovial WBCs ≥ 1,983 cells/μl (80%, 85%), percentage of synovial neutrophils ≥ 76% (80%, 81%), and ANC ≥ 1,684 cells/μl (80%, 87%). Conclusion. Early PJI after TKA should be suspected within six weeks if the CRP is ≥ 82 mg/l, synovial WBCs are ≥ 8,676 cells/μl, the percentage of synovial neutrophils is ≥ 88%, and/or the ANC is ≥ 8,346 cells/μl. Between six and 12 weeks, thresholds include a CRP of ≥ 34 mg/l, synovial WBC of ≥ 1,983 cells/μl, a percentage of synovial neutrophils of ≥ 76%, and/or an ANC of ≥ 1,684 cells/μl. Cite this article: Bone Joint J 2021;103-B(6 Supple A):177–184

Aims. This study used an artificial neural network (ANN) model to determine the most important pre- and perioperative variables to predict same-day discharge in patients undergoing total knee arthroplasty (TKA). Methods. Data for this study were collected from the National Surgery Quality Improvement Program (NSQIP) database from the year 2018. Patients who received a primary, elective, unilateral TKA with a diagnosis of primary osteoarthritis were included. Demographic, preoperative, and intraoperative variables were analyzed. The ANN model was compared to a logistic regression model, which is a conventional machine-learning algorithm. Variables collected from 28,742 patients were analyzed based on their contribution to hospital length of stay. Results. The predictability of the ANN model, area under the curve (AUC) = 0.801, was similar to the logistic regression model (AUC = 0.796) and identified certain variables as important factors to predict same-day discharge. The ten most important factors favouring same-day discharge in the ANN model include preoperative sodium, preoperative international normalized ratio, BMI, age, anaesthesia type, operating time, dyspnoea status, functional status, race, anaemia status, and chronic obstructive pulmonary disease (COPD). Six of these variables were also found to be significant on logistic regression analysis. Conclusion. Both ANN modelling and logistic regression analysis revealed clinically important factors in predicting patients who can undergo safely undergo same-day discharge from an outpatient TKA. The ANN model provides a beneficial approach to help determine which perioperative factors can predict same-day discharge as of 2018 perioperative recovery protocols. Cite this article: Bone Joint J 2021;103-B(8):1358–1366

Aims

The aim of this study was to describe the prevalence and patterns of neuropathic pain over one year in a cohort of patients with chronic post-surgical pain at three months following total knee arthroplasty (TKA).

Aims

A fracture of the medial tibial plateau is a serious complication of Oxford mobile-bearing unicompartmental knee arthroplasty (OUKA). The risk of these fractures is reportedly lower when using components with a longer keel-cortex distance (KCDs). The aim of this study was to examine how slight varus placement of the tibial component might affect the KCDs, and the rate of tibial plateau fracture, in a clinical setting.

Aims. The aim of this study was to evaluate the ability of a machine-learning algorithm to diagnose prosthetic loosening from preoperative radiographs and to investigate the inputs that might improve its performance. Methods. A group of 697 patients underwent a first-time revision of a total hip (THA) or total knee arthroplasty (TKA) at our institution between 2012 and 2018. Preoperative anteroposterior (AP) and lateral radiographs, and historical and comorbidity information were collected from their electronic records. Each patient was defined as having loose or fixed components based on the operation notes. We trained a series of convolutional neural network (CNN) models to predict a diagnosis of loosening at the time of surgery from the preoperative radiographs. We then added historical data about the patients to the best performing model to create a final model and tested it on an independent dataset. Results. The convolutional neural network we built performed well when detecting loosening from radiographs alone. The first model built de novo with only the radiological image as input had an accuracy of 70%. The final model, which was built by fine-tuning a publicly available model named DenseNet, combining the AP and lateral radiographs, and incorporating information from the patient’s history, had an accuracy, sensitivity, and specificity of 88.3%, 70.2%, and 95.6% on the independent test dataset. It performed better for cases of revision THA with an accuracy of 90.1%, than for cases of revision TKA with an accuracy of 85.8%. Conclusion. This study showed that machine learning can detect prosthetic loosening from radiographs. Its accuracy is enhanced when using highly trained public algorithms, and when adding clinical data to the algorithm. While this algorithm may not be sufficient in its present state of development as a standalone metric of loosening, it is currently a useful augment for clinical decision making. Cite this article: Bone Joint J 2020;102-B(6 Supple A):101–106

Aims

Loosening of components after total knee arthroplasty (TKA) can be associated with the development of radiolucent lines (RLLs). The aim of this study was to assess the rate of formation of RLLs in the cemented original design of the ATTUNE TKA and their relationship to loosening.

Aims

We have previously reported the mid-term outcomes of revision total knee arthroplasty (TKA) for flexion instability. At a mean of four years, there were no re-revisions for instability. The aim of this study was to report the implant survivorship and clinical and radiological outcomes of the same cohort of of patients at a mean follow-up of ten years.